BARI is a randomized trial of coronary angioplasty vs bypass surgery for patients with multivessel coronary artery disease requiring revascularization. The study randomized a total of 1,829 patients. In the five-year follow-up, the study showed that 5 year survival was not statistically significantly different between bypass surgery and angioplasty, but angioplasty assigned patients required significantly more procedures (repeat catheterization, angioplasty, and bypass surgery) during the 5-year follow-up. The subset of patients with diabetes mellitus treated with medications at study entry had significantly worse survival when initially assigned to angioplasty than bypass surgery (35% vs 19%, p=.002). Among patients without diabetes requiring treatment at entry, survival was identical in angioplasty and surgical-assigned patients (9% vs 9%). Extended follow-up of the BARI patients is complete.